Server Merge Strategy

PherPhur

AoC seems to already be set up very well for server resets considering there are so many variables in how the world develops both physically and socially. Server resets on a game like this could very well be one of the most novel, interesting and fun aspects about it, depending how how the progression is tuned.

In the AMA we heard Steven talk about one strategy they're looking at for maintaining healthy server population is merging small servers with larger ones and doing a reset on both of them as far as land ownership goes.

So here's my idea in the case that AoC isn't tuned for resets(the progression is slow enough that the majority of players will not have accomplished what they came for and quit playing well before new major content comes out):

There could very well never even need to be server merges. If they just lock and keep locked full servers(WoW Classic is doing this along with bans and it's the best thing any MMO has ever done to combat botting), as well as lock all or most other servers temporarily, while opening up free transfers from all the full servers over to the smallest server until that server is between the new lowest server and the one above it.

They'd essentially just be balancing the servers out. Here's what i've seen, MMOs either seem to belly flop, or they die at a steady pace. If AoC belly flops it's screwed either way, but if it hits it's mark, then it's population will become steady pretty shortly after launch. And people dealt with 3 hour ques in WoW Classic without too much complaining, so I think they should push for longer que times to prevent against feeling the need to do full on merges(if theirs too many servers made at launch during the hype wave then when they try to balance the servers out they'll all be too low in population)

They should only open up the minimum amount of servers they think they'll need, then add a server not when one is full and has a que, but has reached the limit of what's acceptable for the que(things i'm sure they'll already be doing). We heard they will have a limit on the server for "active players", but also one for "assigned players"(players who have a character there), but I think there shouldn't be a limit on assigned players, they should just lock the server the instant the que timer hits what's considered acceptable.

If they do need to do a server merge then the best time to do it would be right after that hype wave die off(likely around 2-3 weeks in). It just so happens virtually no one will have freeholds yet by this time so what they could do is take servers that are deemed too small(Steven called them "weak" servers) and find any other(but not full) servers where they can seamlessly merge them together. What I mean is servers that have no node activity where the other has a developing node(or no activity).

And for the few people that could possibly have freeholds, most likely those can be moved to the exact same spot in that zone as well without overlapping somebody else's structure. Or even being close for that matter.

For the servers that can't be merged this way they can merge 2 or more servers together in a fresh server that has either been completely reset or just has all of it's land reset.

They'd only have to do that during the die-off of the hype wave, only as much until they can remove the "weak" servers. Then they can do what I stated above, which is as far as I see it, a nearly flawless system, a system for maintaining populations to prevent any from becoming a weak server after the hype die-off.

Eventually the population will likely decline to a point where there needs to be another merge, but until then they can just do what I first suggested with the server balances and they won't really have to deal with the die-off from much hype. The hype die-off from new expansions(or major content updates) will be small enough that the que time buffer will eat the blunt of it, hopefully.

Voxtrium

Q times vary, so you would really just balance the Q time during peak hours 8-11pm. I think the population flux is greater than you account for however which negates a lot of this I think.